The most commonly used model in constructed wetlands is the first-order removal model, and first order kinetic constants (k) are the key parameters. The presumption is often made that k are constants. However, it is possible that k are functions of operating conditions, but the influence of operation conditions on k is unclear. In this study, response surface methodology was used to explore the variation patterns of ka (area rate constants) and kV (volume rate constants) for the removal of total nitrogen (TN) and total phosphorus (TP) in free water surface (FWS) wetlands. The experimental variables included hydraulic loading rate (HLR), water depth, and inlet concentration (Cin). The results showed that kV was more variable than ka, and the area-based first-order model is more suitable for simulating TN and TP in FWS wetlands. Inlet concentration (Cin) was significant for ka; Cin and water depth were significant for kV; HLR and the interaction between factors were insignificant. The effects of Cin on ka and kV can be described by an upward convex quadratic curve, while the effect of water depth on kV demonstrates a downward convex quadratic curve. The first-order area rate constant for TN removal was given by k = -47.66 + 22.01 Cin - 1.154 Cin2; the first-order area rate constant for TP removal was given by k = -27.75 + 95.88 Cin - 30.73 Cin2. Based on the variation patterns, the traditional k-C model was modified to the kψ-C model. The kψ-C model produced the best results at simulating the outlet concentration and removal efficiency (RE).
Keywords: First order kinetic constants; First-order removal model; Free water surface wetlands; Response surface methodology; Total nitrogen; Total phosphorus.
Copyright © 2023 Elsevier Ltd. All rights reserved.